CN116110508A - SO of coal-fired unit 2 Method, system, device and medium for early warning of exceeding standard - Google Patents

SO of coal-fired unit 2 Method, system, device and medium for early warning of exceeding standard Download PDF

Info

Publication number
CN116110508A
CN116110508A CN202310016218.5A CN202310016218A CN116110508A CN 116110508 A CN116110508 A CN 116110508A CN 202310016218 A CN202310016218 A CN 202310016218A CN 116110508 A CN116110508 A CN 116110508A
Authority
CN
China
Prior art keywords
concentration
probability
flue gas
superscalar
coal
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202310016218.5A
Other languages
Chinese (zh)
Inventor
钟孝蛟
张磊
朱锋
张世磊
柏恩慈
开乐
徐颋
王照阳
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing Huadun Power Information Security Evaluation Co Ltd
Original Assignee
Nanjing Huadun Power Information Security Evaluation Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nanjing Huadun Power Information Security Evaluation Co Ltd filed Critical Nanjing Huadun Power Information Security Evaluation Co Ltd
Priority to CN202310016218.5A priority Critical patent/CN116110508A/en
Publication of CN116110508A publication Critical patent/CN116110508A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
    • B01D—SEPARATION
    • B01D53/00—Separation of gases or vapours; Recovering vapours of volatile solvents from gases; Chemical or biological purification of waste gases, e.g. engine exhaust gases, smoke, fumes, flue gases, aerosols
    • B01D53/34—Chemical or biological purification of waste gases
    • B01D53/74—General processes for purification of waste gases; Apparatus or devices specially adapted therefor
    • B01D53/80—Semi-solid phase processes, i.e. by using slurries
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
    • B01D—SEPARATION
    • B01D53/00—Separation of gases or vapours; Recovering vapours of volatile solvents from gases; Chemical or biological purification of waste gases, e.g. engine exhaust gases, smoke, fumes, flue gases, aerosols
    • B01D53/34—Chemical or biological purification of waste gases
    • B01D53/46—Removing components of defined structure
    • B01D53/48—Sulfur compounds
    • B01D53/50—Sulfur oxides
    • B01D53/501—Sulfur oxides by treating the gases with a solution or a suspension of an alkali or earth-alkali or ammonium compound
    • B01D53/502—Sulfur oxides by treating the gases with a solution or a suspension of an alkali or earth-alkali or ammonium compound characterised by a specific solution or suspension
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
    • F04B—POSITIVE-DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS
    • F04B15/00—Pumps adapted to handle specific fluids, e.g. by selection of specific materials for pumps or pump parts
    • F04B15/04—Pumps adapted to handle specific fluids, e.g. by selection of specific materials for pumps or pump parts the fluids being hot or corrosive
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
    • F04B—POSITIVE-DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS
    • F04B17/00—Pumps characterised by combination with, or adaptation to, specific driving engines or motors
    • F04B17/03—Pumps characterised by combination with, or adaptation to, specific driving engines or motors driven by electric motors
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
    • F04B—POSITIVE-DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS
    • F04B49/00—Control, e.g. of pump delivery, or pump pressure of, or safety measures for, machines, pumps, or pumping installations, not otherwise provided for, or of interest apart from, groups F04B1/00 - F04B47/00
    • F04B49/06—Control using electricity
    • F04B49/065—Control using electricity and making use of computers
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
    • G08B21/02—Alarms for ensuring the safety of persons
    • G08B21/12—Alarms for ensuring the safety of persons responsive to undesired emission of substances, e.g. pollution alarms
    • G08B21/14—Toxic gas alarms
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
    • G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
    • G16C20/20—Identification of molecular entities, parts thereof or of chemical compositions
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
    • G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
    • G16C20/70—Machine learning, data mining or chemometrics
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
    • B01D—SEPARATION
    • B01D2258/00—Sources of waste gases
    • B01D2258/02—Other waste gases
    • B01D2258/0283—Flue gases

Abstract

The invention discloses a coal-fired unit SO 2 A method, a system, a device and a medium for early warning of exceeding standard, wherein the method comprises the following steps: acquiring historical operation data of a unit; preprocessing historical operation data of the unit to obtain SO 2 A superscalar probability sample library; according to SO 2 Superscalar probability sample library and SO construction 2 A superscalar probability prediction model; according to SO 2 Superscalar probability prediction model for calculating SO of each future period 2 The probability exceeds the standard and gives out early warning; according to the future time periods SO 2 And the variation trend of the exceeding probability along with the total current of the slurry circulating pump gives out the current adjustment suggestion of the slurry circulating pump in each period in the future. The invention comprehensively considers SO caused by various uncertain factors 2 The method has the advantages that the risk exceeding is avoided, the prediction precision is high, and operators can be helped to adjust the desulfurization operation working condition to a more reasonable working condition, so that the economic loss caused by overlarge output of the slurry circulating pump is avoided.

Description

SO of coal-fired unit 2 Method, system, device and medium for early warning of exceeding standard
Technical Field
The invention belongs to the technical field of environmental protection, and in particular relates to a coal-fired unit SO 2 Method, system, device and medium for early warning of exceeding standard.
Background
At present, the energy installation structure in China still takes thermal power as the main material, a great amount of flue gas can be discharged in the operation process of the thermal power plant, and SO exists in the flue gas 2 、NO X And pollutants such as smoke dust, along with the rapid development of economy, the requirements on the treatment of atmospheric pollution are more and more strict, and the task of controlling the pollutant discharge is more and more serious. Wherein SO 2 Is a key requirement for removing pollutants in a thermal power plant.
At present, SO is removed from coal-fired power plants 2 The process is mainly wet desulfurization (FGD), the principle of FGD is that limestone slurry in an absorption tower reacts with sulfur dioxide and the like to generate calcium sulfite, the calcium sulfite reacts with oxygen to generate gypsum, and SO in flue gas in the process 2 The content is reduced. After the coal-fired power plant is transformed by ultralow emission, the flue gas SO is purified 2 Is 35mg/Nm 3 Thus outlet SO 2 The concentration adjustment range is smaller, and the conditions of load fluctuation, coal quantity change, coal blending and combustion and the like of the thermal power plant frequently occur, SO that the raw flue gas SO is caused 2 The concentration greatly fluctuates, which increases a large risk for environmental protection.
The slurry circulation pump is the main power consumption equipment in the FGD system of the thermal power plant. At present, the operation mode of the slurry circulating pump in the domestic thermal power plant desulfurization system is mainly controlled by an operator manually. In order to ensure that the concentration and emission of the sulfur dioxide at the outlet meet the environmental protection requirements, operators generally increase the desulfurization efficiency by increasing the output of a slurry circulating pump and reduce SO (sulfur dioxide) caused by various uncertain factors 2 And (5) risk of exceeding the standard. However, too large slurry circulation pump-out force can increase desulfurization electricity consumption, bring about the increase of station service electricity rate, influence the economy of a unit, lead to adjustment lag due to untimely judgment of operators, and also increase SO 2 And (5) risk of exceeding the standard.
SO of the existing thermal power plant 2 The method for early warning the exceeding of the standard mainly comprises the steps of analyzing single factor pair SO 2 The influence of exceeding the standard, such as the coal feeding amount of the coal feeder, the sulfur content of the fire coal and the like, when the operation data of the factor exceeds the limit value, SO is emitted 2 And (5) exceeding the standard and early warning. In addition, there is also the construction of SO from historical operational data 2 SO for future time period by concentration prediction model 2 A method for predicting concentration.
In the actual operation process of the coal-fired unit, SO of raw flue gas 2 The concentration is influenced by various uncertain factors such as load fluctuation and combustion mode of a unit besides the influence of coal feeding amount and sulfur content of fire coal, and SO caused by various factors is not comprehensively considered in the prior art 2 SO due to fluctuation of concentration 2 And (5) risk of exceeding the standard. At the same time according to SO 2 Prediction model predicts future period SO 2 The prediction result of the concentration has the problem of precisionFailing to quantitatively evaluate the future period SO 2 Is not limited.
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provides a coal-fired unit SO 2 Method, system, device and medium for early warning of exceeding standard, comprehensively considering SO caused by various uncertain factors 2 The method has the advantages that the risk exceeding is avoided, the prediction precision is high, and operators can be helped to adjust the desulfurization operation working condition to a more reasonable working condition, so that the economic loss caused by overlarge output of the slurry circulating pump is avoided.
The invention provides the following technical scheme:
in a first aspect, a coal-fired unit SO is provided 2 The method for warning the exceeding of the standard comprises the following steps:
acquiring historical operation data of a unit;
preprocessing historical operation data of the unit to obtain SO 2 A superscalar probability sample library;
according to SO 2 Superscalar probability sample library and SO construction 2 A superscalar probability prediction model;
according to SO 2 Superscalar probability prediction model for calculating SO of each future period 2 The probability exceeds the standard and gives out early warning;
according to the future time periods SO 2 And the variation trend of the exceeding probability along with the total current of the slurry circulating pump gives out the current adjustment suggestion of the slurry circulating pump in each period in the future.
Further, when historical operation data of the unit are obtained, 30% -100% rated load of the coal-fired unit is taken as a sampling range, and the historical operation data are obtained from an SIS database in a preset sampling time period and sampling period; the acquired unit history operation data comprises: unit load, ambient temperature, sulfur content of fire coal and SO of raw flue gas 2 Concentration, clean flue gas SO 2 Concentration and each slurry circulation pump current.
Further, preprocessing the historical operation data of the unit includes:
calculating SO 2 Removing the sum of the concentration and the current of the slurry circulating pump, removing abnormal data, and forming a sample set;
clustering analysis is carried out on the sample set by taking the total sum of the current of the slurry circulating pumps as a classification standard, then clustering analysis is carried out on each subclass, and SO under the total sum of the current of the slurry circulating pumps of each class is calculated 2 Removing the minimum concentration value;
according to SO 2 Calculating SO of raw flue gas under the sum of current of slurry circulating pumps of each type by removing concentration minimum value 2 A concentration threshold;
under each subclass of the current sum of the slurry circulating pump, carrying out cluster analysis by taking unit load, ambient temperature and sulfur content of fire coal as classification standards to form a plurality of subclass sample data sets;
analyzing each subclass of raw flue gas SO based on subclass sample dataset 2 Concentration distribution rule, and construction of raw flue gas SO 2 A concentration distribution probability density function;
according to the SO of the raw flue gas 2 Concentration threshold and raw flue gas SO 2 Calculating SO of each subclass by using concentration distribution probability density function 2 Superscalar probability, forming SO 2 And (5) a superscalar probability sample library.
Further, the SO 2 The concentration of the removed SO is the original flue gas 2 Concentration and clean flue gas SO 2 The sum of the slurry circulating pump currents is equal to the sum of the slurry circulating pump currents;
the sample set formed includes: unit load, ambient temperature, sulfur content of fire coal and SO of raw flue gas 2 Concentration, clean flue gas SO 2 Concentration, SO 2 The sum of the removal concentration and the slurry circulation pump current.
Further, the current sum of the slurry circulating pumps is used as a classification standard, a K-means clustering algorithm is adopted to classify the sample set, and SO is carried out under the current sum of the slurry circulating pumps of each classification 2 Removing fluctuation of concentration in a certain range, performing cluster analysis on a sample set by adopting a K-means clustering algorithm on each class of the total current of the slurry circulating pump, calculating mass center data of each class, and using SO (SO) 2 Removal of SO from centroid data with minimal concentration 2 The concentration removal value is used as the minimum concentration removal value of SO2 under the total current of the classified slurry circulating pump; SO as to obtain the raw flue gas SO 2 Critical value of concentration is SO 2 Minimum removal concentrationSO 2 And (3) the sum of emission limits.
Further, under each subclass of the current sum of the slurry circulating pump, using unit load, ambient temperature and sulfur content of fire coal as classification standards, adopting a K-means algorithm to perform cluster analysis, and calculating centroid data to obtain the current sum of the slurry circulating pump and SO of raw flue gas 2 Concentration critical value, unit load, ambient temperature and sulfur content of fire coal are all subclasses of classification standards, and sample data of all subclasses comprise: raw flue gas SO 2 Concentration and clean flue gas SO 2 Concentration.
Further, raw flue gas SO 2 The concentration distribution probability density function is:
Figure SMS_1
wherein sigma represents the standard deviation of the SO2 concentration of the raw flue gas, and x represents the SO of the raw flue gas 2 The concentration of the water in the water is higher,
Figure SMS_2
representing the SO of the raw flue gas 2 Mean value of concentration, sigma 2 Representing the SO of the raw flue gas 2 Variance of concentration;
SO 2 the calculation formula of the superscalar probability is as follows:
Figure SMS_3
wherein p is SO 2 Exceeding probability epsilon 0 Representing the SO of the raw flue gas 2 A concentration threshold;
the SO formed 2 The superscalar probability sample library comprises: slurry circulating pump current sum, unit load, environment temperature, sulfur content of fire coal and SO 2 And (5) exceeding the standard probability.
Further, SO 2 The construction method of the superscalar probability prediction model comprises the following steps:
SO based 2 An artificial neural network algorithm is adopted in the superscalar probability sample library, the sum of current of a slurry circulating pump, unit load, ambient temperature and sulfur content of fire coal are taken as input parameters, and S is taken as an input parameterO 2 Training a model built in advance by taking the out-of-standard probability as an output parameter to obtain SO 2 And (5) an out-of-standard probability prediction model.
Further, a future period SO is calculated 2 The method for exceeding the standard probability and sending out the early warning comprises the following steps:
inputting the set load planned value, the current environment temperature, the current coal sulfur content and the current running slurry circulating pump current sum of each future period into SO 2 Calculating and outputting SO of each future period by using an exceeding probability prediction model 2 And sending out early warning of different grades in the range of the exceeding probability.
In a second aspect, a coal-fired unit SO is provided 2 An over-standard warning system comprising:
the data acquisition module is used for acquiring historical operation data of the unit;
the preprocessing module is used for preprocessing the historical operation data of the unit and obtaining SO 2 A superscalar probability sample library;
model building module for according to SO 2 Superscalar probability sample library and SO construction 2 A superscalar probability prediction model;
the calculating and early warning module is used for calculating and early warning according to SO 2 Superscalar probability prediction model for calculating future time period SO 2 And (5) exceeding the probability and giving out early warning.
In a third aspect, a coal-fired unit SO is provided 2 The over-standard warning device comprises a processor and a storage medium, wherein the storage medium is used for storing instructions, and the processor is used for operating according to the instructions to execute the steps of the method in the first aspect.
In a fourth aspect, there is provided a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the method of the first aspect.
Compared with the prior art, the invention has the beneficial effects that:
(1) According to the invention, the historical operation data of the unit is obtained, the historical operation data of the unit is preprocessed, and SO is analyzed 2 Concentration profile, SO 2 The concentration distribution comprehensively reflects the runningBy calculating SO, the influence of various uncertainty factors in the process 2 SO (SO) caused by uncertainty of exceeding probability quantification operation parameters 2 Risk of exceeding standard;
(2) The invention is based on SO 2 SO (SO) construction method for superscalar probability sample library 2 The out-of-standard probability prediction model is combined with a unit load plan to calculate SO of each time period in the future 2 The short-term exceeding early warning is sent out to remind operators of adjusting the desulfurization operation condition in advance, and the prediction accuracy is high;
(3) The invention analyzes SO at various future time intervals 2 The over-standard probability is along with the variation trend of the total current of the slurry circulating pump, the current adjustment suggestion of the slurry circulating pump in each period in the future is given, operators are helped to adjust the desulfurization operation condition to a more reasonable condition, and unnecessary economic loss caused by overlarge output of the slurry circulating pump is avoided.
Drawings
FIG. 1 is a schematic diagram of a coal-fired unit SO in example 1 of the present invention 2 A flow chart of an out-of-standard early warning method;
FIG. 2 is the raw flue gas SO in example 2 of the present invention 2 Concentration profile.
Detailed Description
The invention is further described below with reference to the accompanying drawings. The following examples are only for more clearly illustrating the technical aspects of the present invention, and are not intended to limit the scope of the present invention.
Example 1
As shown in FIG. 1, the embodiment provides a coal-fired unit SO 2 The over-standard early warning method comprises the following steps:
step 1, acquiring historical operation data of a unit.
Step 2, preprocessing historical operation data of the unit to obtain SO 2 And (5) a superscalar probability sample library.
Step 2.1, calculating SO 2 Removing the sum of the concentration and the current of the slurry circulating pump, removing abnormal data, and forming a sample set;
step 2.2, clustering the sample set by taking the total current of the slurry circulating pump as a classification standard, and clustering each subclassAnalysis, calculating SO under the total current of each type of slurry circulating pump 2 Removing the minimum concentration value;
step 2.3 according to SO 2 Calculating SO of raw flue gas under the sum of current of slurry circulating pumps of each type by removing concentration minimum value 2 A concentration threshold;
step 2.4, under each subclass of the total current of the slurry circulating pump, carrying out cluster analysis by taking unit load, ambient temperature and sulfur content of fire coal as classification standards to form a plurality of subclass sample data sets;
step 2.5, analyzing the SO of each subclass of raw flue gas based on the subclass sample data set 2 Concentration distribution rule, and construction of raw flue gas SO 2 A concentration distribution probability density function;
step 2.6, according to the SO of the raw flue gas 2 Concentration threshold and raw flue gas SO 2 Calculating SO of each subclass by using concentration distribution probability density function 2 Superscalar probability, forming SO 2 And (5) a superscalar probability sample library.
Step 3, according to SO 2 The method comprises the steps of constructing an SO (SO) by adopting a neural network algorithm in an out-of-standard probability sample library 2 And (5) an out-of-standard probability prediction model.
Step 4, according to SO 2 Superscalar probability prediction model for calculating SO of each future period 2 And (5) exceeding the probability and giving out early warning.
Step 5, according to the future time periods SO 2 The variation trend of the exceeding probability along with the current sum of the slurry circulating pump is determined SO 2 And when the exceeding probability is reduced to 0, the total current of the slurry circulating pump is reduced, and the current adjustment suggestion of the slurry circulating pump in each period in the future is given.
Example 2
The embodiment provides a coal-fired unit SO 2 The over-standard early warning method comprises the following steps:
step 1, taking 30% -100% rated load of a coal-fired unit as a sampling range, and acquiring historical operation data of the unit from an SIS database according to a preset sampling time period and a preset sampling period; the acquired unit history operation data comprises: unit load, ambient temperature, sulfur content of fire coal and SO of raw flue gas 2 Concentration, clean flue gas SO 2 Concentration and each slurry circulation pumpA current.
Wherein, the raw flue gas SO 2 Concentration refers to SO at inlet of wet desulfurization equipment of coal-fired unit 2 Concentration, i.e. SO 2 Removing SO in the flue gas before 2 Concentration; clean flue gas SO 2 SO with concentration of outlet of wet desulfurization equipment of coal-fired unit 2 Concentration, i.e. SO 2 SO in the flue gas after removal 2 Concentration.
Step 2, calculating SO of each group of data 2 Removing the sum of the concentration and the slurry circulation pump current to form a data set, comprising: unit load, ambient temperature, sulfur content of fire coal and SO of raw flue gas 2 Concentration, clean flue gas SO 2 Concentration, SO 2 The sum of the removal concentration and the current of the slurry circulating pump. Wherein SO 2 Concentration = raw flue gas SO removal 2 Concentration-clean flue gas SO 2 The sum of the concentration and the liquid circulating pump current is the sum of the slurry circulating pump currents.
And 3, eliminating abnormal data in the data set in the step 2 to form a sample set. The rejected outlier data includes:
(1) samples in which any one of the parameters in the dataset is negative;
(2) respectively using a box diagram method to perform the method on the SO of the raw flue gas 2 The outliers of the sum of concentration and slurry circulation pump current were screened and the set of samples was removed.
Step 4, classifying the sample set by adopting a K-means clustering algorithm by taking the current sum of the slurry circulating pumps as a classification standard, and under the current sum of the slurry circulating pumps of each classification, SO 2 Removing fluctuation of concentration in a certain range, performing cluster analysis on a sample set by adopting a K-means clustering algorithm on each class of the total current of the slurry circulating pump, calculating mass center data of each class, and using SO (SO) 2 Removal of SO from centroid data with minimal concentration 2 Removing concentration value as SO under the total current of circulating pump of the classified slurry 2 Minimum removal concentration.
Step 5, according to the result of step 4, SO exists under the total current output of each classified slurry circulating pump 2 Removing the minimum concentration value and purifying the flue gas SO 2 The concentration of the wastewater to reach the standard is required to be smaller than the SO specified in the state 2 Emission limits. According to SO 2 The method for calculating the removal concentration is defined as follows:
raw flue gas SO 2 Concentration critical value=so 2 Minimum removal concentration +SO 2 Emission limits.
When the raw flue gas SO 2 The concentration is greater than that of SO in the original flue gas 2 At the critical value of concentration, the slurry circulating pump current sum output is used for purifying the flue gas SO 2 There is a possibility that the concentration will exceed the standard.
Step 6, under each subclass of the current sum of the slurry circulating pump, using the unit load, the ambient temperature and the sulfur content of the fire coal as classification standards, adopting a K-means algorithm to perform cluster analysis, and calculating mass center data to obtain the current sum of the slurry circulating pump and the SO of the raw flue gas 2 Concentration critical value, unit load, ambient temperature and sulfur content of fire coal are all subclasses of classification standards, and sample data of all subclasses comprise: raw flue gas SO 2 Concentration and clean flue gas SO 2 Concentration.
Step 7, analyzing the raw flue gas SO of each subclass 2 The concentration distribution rule is that the SO2 concentration distribution diagram of the raw flue gas is drawn, as shown in figure 2 2 The concentration profile is generally subject to a normal profile.
Calculating SO of raw flue gas 2 Mean value of concentration
Figure SMS_4
The formula is as follows:
Figure SMS_5
wherein x is i Represents the ith raw flue gas SO 2 The concentration, n, represents the number of the SO2 concentration of the original flue gas;
calculating SO of raw flue gas 2 Variance sigma of concentration 2 The formula is as follows:
Figure SMS_6
due to the SO of raw flue gas 2 The concentration is subject to normal distribution, SO that the raw flue gas SO 2 The concentration distribution probability density function is:
Figure SMS_7
wherein sigma represents the standard deviation of the SO2 concentration of the raw flue gas, and x represents the SO of the raw flue gas 2 Concentration.
Step 8, using the sum of current of slurry circulating pumps and SO of raw flue gas 2 The concentration critical value, the unit load, the ambient temperature and the sulfur content of the fire coal represent different operation conditions for each subclass of the classification standard, and the SO of the operation condition corresponding to each subclass 2 The exceeding probability is the SO of the original flue gas 2 The concentration is greater than that of SO in the original flue gas 2 Probability of concentration threshold. SO (SO) 2 The calculation formula of the superscalar probability p is as follows:
Figure SMS_8
wherein ε 0 Representing the SO of the raw flue gas 2 Concentration threshold.
Based on the SO from each subclass 2 Superscalar probability, taking each subclass as a sample to form new SO with different working conditions 2 A superscalar probability sample library comprising: slurry circulating pump current sum, unit load, environment temperature, sulfur content of fire coal and SO 2 And (5) exceeding the standard probability.
Step 9, based on the SO obtained in step 8 2 The superscalar probability sample library adopts an artificial neural network algorithm, takes the sum of current of a slurry circulating pump, unit load, ambient temperature and sulfur content of fire coal as input parameters, and takes SO as input parameters 2 Training a model built in advance by taking the out-of-standard probability as an output parameter to obtain SO 2 And (5) an out-of-standard probability prediction model.
Step 10, applying SO 2 Superscalar probability prediction model for unit SO for short periods of time, e.g., 15 minutes, 30 minutes, 1 hour, 2 hours, in the future 2 And (5) early warning is carried out under the condition of exceeding the standard. Specifically, the unit load planned value, the current ambient temperature, the current coal sulfur content and the current running slurry circulating pump current sum of all future time periods are input into the step9 SO obtained 2 In the superscalar probability prediction model, SO in each future period is output after calculation 2 And sending out early warning of different grades in the range of the exceeding probability.
Step 11, analyzing the future time periods SO obtained in step 10 2 The variation trend of the exceeding probability along with the current sum of the slurry circulating pump is determined SO 2 And when the exceeding probability is reduced to 0, the total current of the slurry circulating pump is reduced, and the current adjustment suggestion of the slurry circulating pump in each period in the future is given.
Example 3
The embodiment provides a coal-fired unit SO 2 The method of embodiment 1 or 2 is adopted in the standard exceeding early warning system, and specifically includes:
the data acquisition module is used for acquiring historical operation data of the unit;
the preprocessing module is used for preprocessing the historical operation data of the unit and obtaining SO 2 A superscalar probability sample library;
model building module for according to SO 2 Superscalar probability sample library and SO construction 2 A superscalar probability prediction model;
the calculating and early warning module is used for calculating and early warning according to SO 2 Superscalar probability prediction model for calculating future time period SO 2 And (5) exceeding the probability and giving out early warning.
Example 4
The embodiment provides a coal-fired unit SO 2 The over-standard warning device comprises a processor and a storage medium, wherein the storage medium is used for storing instructions, and the processor is used for operating according to the instructions to execute the steps of the method in the embodiment 1 or 2.
Example 5
The present embodiment provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method described in embodiment 1 or 2.
The foregoing is merely a preferred embodiment of the present invention, and it should be noted that modifications and variations could be made by those skilled in the art without departing from the technical principles of the present invention, and such modifications and variations should also be regarded as being within the scope of the invention.

Claims (12)

1. SO of coal-fired unit 2 The method for warning the exceeding of the standard is characterized by comprising the following steps:
acquiring historical operation data of a unit;
preprocessing historical operation data of the unit to obtain SO 2 A superscalar probability sample library;
according to SO 2 Superscalar probability sample library and SO construction 2 A superscalar probability prediction model;
according to SO 2 Superscalar probability prediction model for calculating SO of each future period 2 The probability exceeds the standard and gives out early warning;
according to the future time periods SO 2 And the variation trend of the exceeding probability along with the total current of the slurry circulating pump gives out the current adjustment suggestion of the slurry circulating pump in each period in the future.
2. The coal-fired unit SO of claim 1 2 The method is characterized in that when historical operation data of a unit are acquired, 30% -100% rated load of a coal-fired unit is taken as a sampling range, and a preset sampling time period and a preset sampling period are acquired from an SIS database; the acquired unit history operation data comprises: unit load, ambient temperature, sulfur content of fire coal and SO of raw flue gas 2 Concentration, clean flue gas SO 2 Concentration and each slurry circulation pump current.
3. The coal-fired unit SO of claim 1 2 The out-of-standard early warning method is characterized by comprising the following steps of:
calculating SO 2 Removing the sum of the concentration and the current of the slurry circulating pump, removing abnormal data, and forming a sample set;
clustering analysis is carried out on the sample set by taking the total sum of the current of the slurry circulating pumps as a classification standard, then clustering analysis is carried out on each subclass, and SO under the total sum of the current of the slurry circulating pumps of each class is calculated 2 Removing the minimum concentration value;
according to SO 2 Calculating SO of raw flue gas under the sum of current of slurry circulating pumps of each type by removing concentration minimum value 2 A concentration threshold;
under each subclass of the current sum of the slurry circulating pump, carrying out cluster analysis by taking unit load, ambient temperature and sulfur content of fire coal as classification standards to form a plurality of subclass sample data sets;
analyzing each subclass of raw flue gas SO based on subclass sample dataset 2 Concentration distribution rule, and construction of raw flue gas SO 2 A concentration distribution probability density function;
according to the SO of the raw flue gas 2 Concentration threshold and raw flue gas SO 2 Calculating SO of each subclass by using concentration distribution probability density function 2 Superscalar probability, forming SO 2 And (5) a superscalar probability sample library.
4. The coal-fired unit SO of claim 3 2 A method for early warning of exceeding standard is characterized in that the SO 2 The concentration of the removed SO is the original flue gas 2 Concentration and clean flue gas SO 2 The sum of the slurry circulating pump currents is equal to the sum of the slurry circulating pump currents;
the sample set formed includes: unit load, ambient temperature, sulfur content of fire coal and SO of raw flue gas 2 Concentration, clean flue gas SO 2 Concentration, SO 2 The sum of the removal concentration and the slurry circulation pump current.
5. The coal-fired unit SO of claim 3 2 The over-standard early warning method is characterized in that the total current of the slurry circulating pump is used as a classification standard, a K-means clustering algorithm is adopted to classify a sample set, and SO is carried out under the total current of the slurry circulating pump of each classification 2 Removing fluctuation of concentration in a certain range, performing cluster analysis on a sample set by adopting a K-means clustering algorithm on each class of the total current of the slurry circulating pump, calculating mass center data of each class, and using SO (SO) 2 Removal of SO from centroid data with minimal concentration 2 The concentration removal value is used as the minimum concentration removal value of SO2 under the total current of the classified slurry circulating pump; SO as to obtain the raw flue gas SO 2 Critical value of concentration is SO 2 Minimum concentration of removal and SO 2 And (3) the sum of emission limits.
6. The coal-fired unit SO of claim 3 2 A method for early warning of excessive standard is characterized in that under each subclass of the total current of a slurry circulating pump, using unit load, ambient temperature and sulfur content of fire coal as classification standards, adopting K-means algorithm to perform cluster analysis, calculating mass center data and obtaining the total current of the slurry circulating pump and SO of raw flue gas 2 Concentration critical value, unit load, ambient temperature and sulfur content of fire coal are all subclasses of classification standards, and sample data of all subclasses comprise: raw flue gas SO 2 Concentration and clean flue gas SO 2 Concentration.
7. The coal-fired unit SO of claim 3 2 The over-standard early warning method is characterized in that the original flue gas SO 2 The concentration distribution probability density function is:
Figure FDA0004040462300000031
wherein sigma represents the standard deviation of the SO2 concentration of the raw flue gas, and x represents the SO of the raw flue gas 2 The concentration of the water in the water is higher,
Figure FDA0004040462300000032
representing the SO of the raw flue gas 2 Mean value of concentration, sigma 2 Representing the SO of the raw flue gas 2 Variance of concentration;
SO 2 the calculation formula of the superscalar probability is as follows:
Figure FDA0004040462300000033
wherein p is SO 2 Exceeding probability epsilon 0 Representing the SO of the raw flue gas 2 A concentration threshold;
the SO formed 2 The superscalar probability sample library comprises: slurry circulating pump current sum, unit load, environment temperature, sulfur content of fire coal and SO 2 And (5) exceeding the standard probability.
8. The coal-fired unit SO of claim 1 2 A method for early warning of exceeding standard is characterized in that 2 The construction method of the superscalar probability prediction model comprises the following steps:
SO based 2 The superscalar probability sample library adopts an artificial neural network algorithm, takes the sum of current of a slurry circulating pump, unit load, ambient temperature and sulfur content of fire coal as input parameters, and takes SO as input parameters 2 Training a model built in advance by taking the out-of-standard probability as an output parameter to obtain SO 2 And (5) an out-of-standard probability prediction model.
9. The coal-fired unit SO of claim 1 2 The over-standard early warning method is characterized by calculating a future period SO 2 The method for exceeding the standard probability and sending out the early warning comprises the following steps:
inputting the set load planned value, the current environment temperature, the current coal sulfur content and the current running slurry circulating pump current sum of each future period into SO 2 Calculating and outputting SO of each future period by using an exceeding probability prediction model 2 And sending out early warning of different grades in the range of the exceeding probability.
10. SO of coal-fired unit 2 The utility model provides a superscalar early warning system which characterized in that includes:
the data acquisition module is used for acquiring historical operation data of the unit;
the preprocessing module is used for preprocessing the historical operation data of the unit and obtaining SO 2 A superscalar probability sample library;
model building module for according to SO 2 Superscalar probability sample library and SO construction 2 A superscalar probability prediction model;
the calculating and early warning module is used for calculating and early warning according to SO 2 Superscalar probability prediction model for calculating future time period SO 2 And (5) exceeding the probability and giving out early warning.
11. SO of coal-fired unit 2 Super standard early warning device, its specialCharacterized by comprising a processor and a storage medium for storing instructions, the processor being operative according to the instructions to perform the steps of the method of any one of claims 1 to 9.
12. A computer readable storage medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, implements the steps of the method according to any one of claims 1-9.
CN202310016218.5A 2023-01-06 2023-01-06 SO of coal-fired unit 2 Method, system, device and medium for early warning of exceeding standard Pending CN116110508A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310016218.5A CN116110508A (en) 2023-01-06 2023-01-06 SO of coal-fired unit 2 Method, system, device and medium for early warning of exceeding standard

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310016218.5A CN116110508A (en) 2023-01-06 2023-01-06 SO of coal-fired unit 2 Method, system, device and medium for early warning of exceeding standard

Publications (1)

Publication Number Publication Date
CN116110508A true CN116110508A (en) 2023-05-12

Family

ID=86266864

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310016218.5A Pending CN116110508A (en) 2023-01-06 2023-01-06 SO of coal-fired unit 2 Method, system, device and medium for early warning of exceeding standard

Country Status (1)

Country Link
CN (1) CN116110508A (en)

Similar Documents

Publication Publication Date Title
CN109493250B (en) Method for evaluating denitration capability of SCR reactor
CN111077869B (en) Big data intelligent control bag-type dust collector optimization control method and system
CN111240282B (en) Process optimization method, device, equipment and computer readable storage medium
CN112465239B (en) Desulfurization system operation optimization method based on improved PSO-FCM algorithm
Qiao et al. An investigation on data mining and operating optimization for wet flue gas desulfurization systems
CN110889088A (en) Enterprise pollution discharge supervision method assisted by electric model
CN108448625B (en) Fan cut-in wind speed calculation method based on data driving
CN114749006A (en) Method for optimizing wet desulphurization slurry circulating pump
CN111522290A (en) Denitration control method and system based on deep learning method
CN113792988A (en) Online monitoring data anomaly identification method for enterprise
CN112364562A (en) Cooperative control method and system for flue gas environmental protection island
CN112232373A (en) Method and system for predicting carbon content of boiler fly ash
CN113440990A (en) EMD-LSTM based outlet SO2Concentration prediction method
CN111598308B (en) Method for solving combination optimization of slurry circulating pump based on regression and double PSO algorithm
CN114632624B (en) Electric dust removal operation optimization system and optimization method
CN117151701B (en) Industrial waste heat recycling system for cogeneration
CN116110508A (en) SO of coal-fired unit 2 Method, system, device and medium for early warning of exceeding standard
CN111564184A (en) Limestone-gypsum wet desulphurization SO of coal-fired power plant3Collaborative removal efficiency prediction method
CN115146871A (en) Intelligent desulfurization system based on data
CN115145152A (en) Boiler combustion and denitration process collaborative optimization control method
CN112835950B (en) System and method for acquiring standard emission operation curve of wet desulphurization system based on DCS data mining
CN112044270B (en) Desulfurization oxidation air system control method based on data-driven multiple models
CN105809304B (en) Method for analyzing correlation of production operation parameters of power plant and pollution control facility
CN112107977A (en) Automatic pH value adjusting method, system and device based on desulfurization system
CN112070303A (en) Parameter-adaptive photovoltaic power ramp event hierarchical probabilistic prediction method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination